Human-AI Symbiosis for Agile Planning

Abstract

The key goal of this proposal is to enhance the synergy of hybrid human-machine teams to become more effective units, achievable via efficient multi-objective optimization algorithms for agile mission planning that provide proactive, context-dependent decision support with enhanced operational capability under uncertainty, time pressure and resource constraints. The proposed decision support capabilities are applicable across a broad range of Navy-relevant missions impacted by uncertainty, such as ship/helicopter/UAV/submarine/aircraft carrier strike group routing, multi-domain battle management, and unmanned system coordination. UCONN-NJIT-NRL- MRY team seeks to investigate (1) adaptive hierarchical causal models and variational free energy to represent, infer and predict context and to address contingencies requiring a rapid response and slower contextual changes that require adaptation over time, (2) Information valuation and prioritization for context-driven operations via information-theoretic and ???wrapper-based what-if??? analysis approaches, (3) Practical interactive multi-objective optimization algorithms for resource allocation (for surveillance and execution) and asset routing, informed by context (mission, environment, asset, threat, human cognition) and data, featuring adaptive search interfaces (e.g., scatter/gather for exploratory search tasks, baseline web search for lookup-type query tasks, Q-learning, multi-grid methods, feature-based aggregation, rollout, deep reinforcement learning and approximate policy iteration), (4) An adaptive, distributed context-aware semi-supervised team decision-making framework featuring bounded rationality under uncertainty concepts, human attributes and collaborative perception and action selection for efficient and effective communication of context-driven high value information dissemination across agents in disconnected, intermittent, and low-bandwidth environments, (5) Evaluation of proactive decision support concepts leveraging different mission simulation testbeds (Multidomain Battlespace Management (MDBM), Common Submarine Coordinated Asset Planner for Engagement (C-SCAPE)) and current decision support tools (COAST, CONFIDENT, TMPLAR).

Document Details

Document Type
DoD Grant Award
Publication Date
Sep 19, 2018
Source ID
N000141812838

Entities

People

  • Krishna R. Pattipati

Organizations

  • Office of Naval Research
  • United States Navy
  • University of Connecticut

Tags

Fields of Study

  • Computer science

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Enterprise Information Systems Architecture and Joint Command Capability Interoperability Support.
  • Operations Research

Technology Areas

  • AI & ML
  • AI & ML - Autonomous Systems
  • AI & ML - DoD AI Strategy
  • AI & ML - Machine Learning Algorithms
  • Autonomy
  • Autonomy - Autonomous System Control
  • Autonomy - Human-Robot Interaction
  • Autonomy - UAVs